What's The "Best" Way To Measure Work in a Cycling Workout?
Earlier this year I started training for the odd ITU distance of 3k/120k/30k. It's not a combo that comes up very often, and in fact in 20 years of triathlon, I'd only done it once before, two years earlier. So instead of having a fairly intuitive sense of how much and what type of work to do, garnered from scores of races and training cycles for all the standard triathlon distances, I was at a loss for what might work best. Coach P got me on track by asking, "So what worked the last time you did this?"
The problem was compounded by the radical difference between the two courses. The first time I did that distance, it was flat and fast, and I went about 3:53; the course this year was totally hilly, and 6 miles shorter. I expected to be out there for up to an hour more. But I didn't want to be any more tired for the run. I turned to kilojoules for help, noting that I'd done 2035 the first time, and so I aimed all my long rides, be they outdoors, inside, hard or easy, to come in at about the same #. The reason I picked that was it was easily available - it's one of the three metrics (the other two are time and distance) which appears on the log page of PowerAgent, the app where I can easily find all of my bike rides.
Turns out, when normalized for the distance difference, I did pretty much the same amount of work: 2000 kj compared to 2035. That got me to thinking, maybe I should be looking at kj more than I do, in addition to all the other metrics: NP, Avg P, TSS, and the cumulative numbers for acute and chronic training load, training stress balance which are calculated from each workout's TSS, which itself is calculated from the ride's NP, which is derived from the AP, which comes from...kilojoules.
There are lots of equations linking all of the above, none of which I can quote, and all of which I guess contain some assumptions about how work accumulates over time and affects the outcome we want: more fitness. I worry all that data manipulation has the potential to wash out, to some degree, the value of the work actually being done during the ride.
Anybody else out there care about kilojoules, or use them in a meaningful way either for training or racing?
Comments
Because no-one else has answered, I will make an observation FWIW.
I don't care about kilojoules for training purposes. For energy balance purposes, yes I do take notice (mainly because my appetite isn't very well calibrated).
In terms of training, I expect kilojoules suffers from the same drawback as TSS in that kilojoules expended well above threshold have way more impact on your muscles and fuel stores than work at IM pace.
@Al Truscott I remember you tracking KJ's at our mini AZ camp, all good points. I kinda like the idea for cycling as the KJ's are irrelevant to what your FTP(on that day) is, or more importantly what your FTP setting is, and could be tracked season to season with no real variables. KJ's are a simple measurement , a none debatable measurement of work performed.
{I worry all that data manipulation has the potential to wash out, to some degree, the value of the work actually being done during the ride.}
I have always felt this way about NP vs. AP and feel NP is overrated or over used... NP is a mathematical estimate of what your watts may have been, or may have felt like , if you had ridden smoother (to me this is like coulda shoulda woulda but didn't)... AP is the real work performed and what actually determines your outcome (speed) or time from point a to point b . Estimates vs. Facts similar to FTP discussions.
Found this interesting short article on TSS vs. Kilojoules , nothing new to what Al said but interesting.
https://wattmatters.blog/home/2006/12/tss-vs-kj.html
@Al Truscott I think it "Works" when you are comparing apples to apples (same distance + strategy to same distance and strategy). You can't take a flat 56 HIM bike KJ # and apply that to Hilly 60 mile bike race.
I personally use TSS as I can get that on my display and I am used to it...
Anyone else? @Rich Stanbaugh ? @John Withrow ?
@tim cronk do you display NP or AP, and same for laps on your devices when training or racing?
KJ is a measure of energy/work but don't know how it relates to watts or the relation between the three. However, I've heard the Trainer Road guys, on their podcast, talk about using KJ numbers to know how many calories to consume post workout to replace the caloric deficit in order to not under or over eat. Something about it being more accurate that calories burned number.
@scott dinhofer I still display NP on my devices. I am always focussed on my 3sec power, When I am doing shorter intervals 5' and under I understand that my AP is probably higher than my NP displayed, when I am doing longer intervals(NP is usually higher than AP) so focussing on NP usually nets me a higher AP. IMO for training its fine to go for highest NP/TSS combo via high VI type riding , but when racing we all know that a lower VI will usually net you the fastest time which is of course your AP. STRAVA is always shown in AP, ZWIFT speed is calculated using AP, TP uses AP to calculate your w/kg for any segment of your ride you are looking at. AP is real, NP is math... WKO4 has charts NP / AP PDC.
@Al Truscott , I’ve thought about this too, trying to figure out how to measure toughness of a course in order to train for it.
Work = watts * time
TSS = intensity ^2 * 100 time
Both of these metrics are driven by how long and how hard you press the pedals. TSS emphasizes / punishes you for pressing harder than the Pavg for the ride. Work doesn’t account for that.
I’ve also looked at CVI, or elevation feet gained / mile as a way to compare courses.
What I’ve been doing recently is different.
1. I model my ride on the course in best bike split.
2. I save the BBS ride as a .tcx file
3. I load the .tcx file into WKO as if it were a ride that I had actually done.
4. I analyze that ride as if i had already ridden the course.
5. Based on 4, I make tweaks to the BBS plan, export the updated plan as a .tcx and repeat the process.
I think of this as a kind of “pre mortem” of the ride plan be a “post mortem” of the actual ride.
This comports with what I;ve learned looking into kilojoules. All of our devices and apps give us a "Calorie" estimate for a workout, which is a calculated number derived from kj. A calorie (small c) is the amount of energy (kj) it takes to raise 1 ml of water 1 deg Cent., or something like that. A Calorie (large C) is the number we see on our devices, simply 1000 calories. There is are two constants which are applied to the kj to get the Calorie. One is, I think, 4.186, representing something I forget; the other is, in Training Peaks' case, 0.22, which is an estimate of an estimate. The human body is inefficient at converting fuel into work, variously estimated at between 0.18 and 0.26 efficient (meaning 74-82% of the calories are lost to heat, I guess), so they use 0.22 as their best guess of a bunch of best guesses.
@Derrek Sanks Joule is a measure of Work. Watt is measure of Power, or rate of doing work.
Watts x seconds = Joules
A kiloJoule is 1,000 Joules
Here is an example from training peaks from my ride today:
Looking at my TP data (below) you see that it says I did 3,272KJ of work.
The difference is only rounding - Pavg is not exactly 193w, it is closer to 193.3 (which I just confirmed in WKO where I can control the number of decimal places I see).
When you calculate the work done with a bike power meter, it is a lot more accurate than something that just estimates a number based off of pace or speed, because the power meter is measuring how hard you press the pedals (unless you have your power meter in your pedals, then it works differently).
Al showed all the math for converting work performed to calories burned - one factor (22%) is the efficiency of your body at converting metabolic energy to mechanical energy and a calorie ≈ 4.18J.
If you work with the assumption that you can roughly replace 50% of the energy you burn while biking with food (energy) you consume, you can use look at TP to see KJ work done, divide by 2 and get a rough estimate for how many calories you should consume on an IM bike. I forget where I read the 50%, but think it is commonly used.
If a watt was a watt was a watt, I’d say tss or kJ. But my experience is that isn’t the case.